/. Embryo/, exp. Morph. Vol. 45, pp. 93-105, 1978
Printed in Great Britain © Company of Biologists Limited 1978
93
In vitro development of inner cell masses isolated
immunosurgically from mouse blastocysts
I. Inner cell masses from 3*5-day p.c. blastocysts
incubated for 24 h before immunosurgery
ByBRlGID HOGAN 1 AND RITA TILLY 1
From the Imperial Cancer Research Fund,
Mill Hill Laboratories, London
SUMMARY
This paper describes the in vitro development of inner cell masses isolated immunosurgically from mouse blastocysts which had been collected on 3-5 days p.c. and then incubated
for 24 h. The inner cell masses continue to grow in culture and develop through a series of
stages with increasing complexity of internal organization. By day 1 all of the cultured
ICMs have an outer layer of endoderm, and by day 3 some of them have two distinct kinds
of inside cells; a columnar epithelial layer and a thin hemisphere of elongated cells. Later,
mesodermal cells appear to delaminate from a limited region of the columnar layer, close
to where it forms a junction with the thinner cells. By day 5, about 25 % of the cultured
ICMs have a striking resemblance to normal 7-5-day p.c. C3H embryos, with embryonic
ectoderm, extra-embryonic ectoderm and chorion, embryonic and extra-embryonic mesoderm,
and visceral endoderm. When mechanically disrupted and grown as attached clumps of cells
in a tissue dish, these embryo-like structures give rise to trophoblast-like giant cells. These
results suggest that the inner cell mass of 4-5-day p.c. blastocysts contains cells which can
give rise to trophoblast derivates in culture.
INTRODUCTION
The preimplantation mouse blastocyst consists of an outer layer of trophectoderm surrounding the blastocoele cavity and inner cell mass. A technique known
as 'immunosurgery' has recently been described (Solter & Knowles, 1975)
for selectively killing the outer trophectoderm cells by exposing the blastocysts
sequentially to antispecies antibody, washing medium and complement.
Following immunosurgery, the isolated inner cells continue to multiply and
differentiate in vitro for up to several weeks. This combination of immunosurgery
and in vitro culture may provide information about the development potential
of the inner cells at different times after blastocyst formation, and might be
used to build up a picture of the timing and interrelationship of cell determination, morphogenesis and cell lineage in the mammalian embryo. In this and the
1
Authors' address: Imperial Cancer Research Fund, Mill Hill Laboratories, Burtonhole
Lane, London, NW7 IAD, U.K.
7
EMB
45
94
B. HOGAN AND R. TILLY
Collect blastocysts
Immunosurgery
24.00 h
I
6
1
I
7
2
I
8
I
9
3
4
5
Days in culture
I
10
^
Daysp.c.
6
7
Fig. 1. Relative timing of normal pregnancy, immunosurgery and in vitro
incubation of embryos.
subsequent paper we describe in detail the growth and morphogenesis of
inner cell masses isolated from C3H/He blastocysts of different ages. For
technical reasons, the initial experiments used blastocysts collected on 3-5 days
p.c. and then incubated in vitro for 24 h before immunosurgery; the development of the inner cell masses from these blastocysts is the subject of this paper.
The development of inner cell masses isolated from blastocysts collected between
3-5 and 4-0 p.c. is described elsewhere (Hogan & Tilly, 1978).
METHODS
Embryos were obtained from 8- to 12-week-old normally mated inbred
C3H/He mice and cultured in vitro according to the schedule shown in Fig. 1.
12.00 h on the day on which plugs were first observed was considered to be
0-5 days post coitum (p.c.)
Isolation of blastocysts
Blastocysts were flushed from uteri between 14.00 and 16.00 h on the 4th day
of pregnancy (approximately 3-6 days p.c). Embryos from the same female
varied in their stage of development; the majority appeared by phase contrast
microscopy to be either substage 1 (early) or substage 2 (mid) blastocysts
according to the classification of Nadijcka & Hillman (1974) (see also fig. 2,
Hogan & Tilly, 1978). The zonae pellucidae were removed with a brief incubation at 37 °C in pronase (0-5 % (w/v) in PBS with 0-5 % polyvinylpyrrolidone) and the blastocysts incubated 24 h in Dulbecco's modified Eagle's
medium (DMEM) containing 20 % human cord serum (HCS, heat inactivated
56 °C for 30 min) in a humidified air/CO 2 incubator at 37 °C.
Immunosurgery
The antiserum used in all experiments was prepared by bleeding a rabbit 10
days after three fortnightly injections of 4 x 108 C57BL spleen cells. The serum
was heat inactivated at 56 °C for 30 min and was found to be effective in the
immunosurgery procedure at dilution of up to 1:200.
In vitro development of inner cell masses. I
95
Table 1. Differentiation in vitro of immunosurgically isolated inner cell masses
into structures of different classes, as judged by phase contrast microscopy
Examples of the three types of structure are shown in Fig. 3. and are described in
the text. In A, which represents the combined results of three separate experiments,
the dead trophectoderm was not removed after immunosurgery. In B, which represents the combined results of two separate experiments, the trophectoderm was removed by sucking the embryos into a finely drawn out Pasteur pipette.
Day 5
Type
Day 3
Type
A
(A) Dead
after
(B) Dead
after
trophectoderm not removed
immunosurgeiy
trophectoderm removed
immunosurgery
1
II
I
II
III
73
35
45
34
29
40
21
26
22
13
Method A
Immunosurgery was routinely carried out between 14.00 and 16.00 h (equivalent to 4-6 days p.c). Blastocysts, transferred in a few microlitres of medium,
were incubated for 30 min at 37 °C in 3 ml of antiserum diluted 1:30 with
DMEM containing 0-5 % bovine serum albumin (Sigma fraction V) or 10 %
heat inactivated foetal calf serum. They were then washed twice in 3 ml of the
same medium and transferred to 3 ml of guinea-pig complement (Wellcome
freeze-dried guinea-pig serum, diluted 1:10). After 30 min at 37 °C the outer
trophectoderm cells had all lysed and the embryos were washed twice in 3 ml
supplemented DMEM. The final incubation was in 3 ml of DMEM plus 20 %
HCS and the medium was changed every 3-4 days.
In contrast to Solter & Knowles (1975) we routinely did not remove the dead
trophectoderm cells before incubating the embryos further. We found that
the inner cell masses grew well inside their trophectoderm shells and were
prevented by them from aggregating together and from attaching to the substratum; after about 20 h the inner cell masses separated from the shells
spontaneously. Control experiments showed that similar results were obtained
when the dead trophectoderm cells were removed by pipetting immediately after
exposure to complement (see Table 1).
Light microscopy and electron microscopy
Embryos were placed in 2-5 % glutaradehyde buffered with 0-06 M sodium
cacodylate at pH 7-3 for 1 h, rinsed in three changes of cacodylate buffer and
post-fixed in 1 % osmium tetroxide buffered with veronal acetate at pH 7-3
Dehydration was carried out in a series of graded ethanols before embedding
in Araldite medium.
For histological examination several sections each 1 /im thick were taken
7-2
96
B. HOGAN AND R. TILLY
Fig. 2. Typical blastocyst at the time of immunosurgery, showing a layer of endoderm cells on the blastocoele side of the inner cell mass. This corresponds to substage
3 of Nadijcka & Hillman (1974). In some blastocysts, endoderm cells had migrated
along the trophectoderm (substage 4). Bar is 25 fim.
every 10-20 /im through the embryos. As the sections were cut on a Cambridge
Ultramicrotome, thin sections could also be taken at these intervals for electron
microscopy. The sections for light microscopy were mounted on glass slides
and stained with 1 % toluidine blue in 1 % borax solution, and the sections for
electron microscopy were mounted on copper grids and stained with uranyl
acetate solution followed by lead citrate, before being examined in a Siemens
Elmskop 1 electron microscope.
RESULTS
Immunosurgery of blastocysts
At the time when immunosurgery was routinely carried out (equivalent to
4 days 14 h p.c.) all of the embryos were fully expanded blastocysts, and in
section had a layer of endoderm cells on the blastocoele side of the inner cell
mass (Fig. 2). They corresponded to substage 3 or 4 blastocysts according to the
classification of Nadijcka & Hillman (1974). Judging from electron microscope
sections, there was no obvious proliferation of polar trophectoderm to form
In vitro development of inner cell masses. I
Fig. 3. Phase-contrast microscopy of isolated inner cell masses at different times
during in vitro incubation: (A) after 20 h, (B) after 48 h (Type I), (C) after 3 days
(Types I and II), and (D) after 5 days (Type III). Bar is 100 /tm in A-C and 200 fim
in D.
97
98
B. HOGAN AND R. TILLY
B
Fig. 4. Sections through typical cultured ICMs at different times after immunosurgery. (A) After 48 h (Type I), (B) after 3 days (Type II), showing the two populations of inside cells; columnar epithelial and elongated. Bar is 50 /*m.
In vitro development of inner cell masses. I
99
ectoplacental cone. In fourteen blastocysts fixed immediately after immunosurgery and serially sectioned at 1 jum, all the outer cells, including those above
the inner cell mass, were clearly lysed, and the nuclear membranes disrupted.
Inner cell masses 20-24 h after immunosurgery
After 20 h in culture, immunosurgically isolated inner cell masses separate
from their trophectodermal shells but remain floating and do not attach to the
substratum (Fig. 3 A). They consist of a solid core of cells surrounded by a
monolayer of endoderm cells which is probably formed by the growth and
spreading of the endoderm cells present before immunosurgery. A thin layer
of finely fibrillar basement membrane material is present between the outer
endoderm cells and the inner core; this basement membrane is presumably
secreted by the overlying endoderm cells which have some rough endoplasmic
reticulum, as well as microvilli, vacuoles and cytoplasmic inclusions.
Cultured inner cell masses 48 h after immunosurgery
After 48 h, most of the cultured ICMs have a central cavity within the epiblast
(embryonic ectoderm) (Fig. 3B, 4A). This central cavity appears to be formed
initially by cell death; in many instances dead cells and debris are present in the
centre, while adjacent cells are completely healthy and have numerous microvilli on their free surface.
The cells of the outer endoderm layer, which in some cases have proliferated
to form loosely attached clumps (Fig. 3B), now clearly resemble visceral endoderm cells of the normal embryo (Solter, Damjanov & Skreb, 1970) and the
basement membrane remains thin.
Cultured inner cell masses 3 days after immunosurgery
After 3 days in culture all of the cultured ICMs have developed into hollow
vesicles, but they show a variable degree of expansion (Fig. 3C). About 67 %
are only slightly larger than at 48 h and are classified as Type I structures in
Table 1. The remainder are considerably more expanded and in the phase
contrast microscope some of the inner cells are distinctly thinner than the rest;
these are classified as Type II structures. As shown in Table 1, the proportion
of the two types present on day 3 was not significantly affected by removing
the dead trophectoderm cells after immunosurgery.
In sections of Type II structures there is clearly a subdivision of the inner
vesicle into two regions; one in which the cells are packed into an epithelial
layer, and another in which the cells are much more elongated (Fig. 4B).
Apart from their shape, there is very little difference in the morphology of the
cells in the two regions and they have few cytoplasmic organelles and little
endoplasmic reticulum. On the side facing the proamniotic cavity the cells are
joined apically by junctional complexes and they have some microvilli which are
covered in a layer of extracellular material. Some of the elongated cells interdigitate along their adjoining surfaces, but this is not a regular feature.
100
B. HOGAN AND R. TILLY
Fig. 5. Section through a typical cultured ICM fixed 5 days after immunosurgery.
The structure has collapsed during subsequent handling. The arrows show the
junction between the columnar epithelial layer and the hemisphere of more rounded
cells. Early mesoderm cells appear to be delaminating from a localized region near
the junctional zone. Bar is 50 /mi.
In vitro development of inner cell masses. I
101
Fig. 6. Sections taken at two levels through a single Type III cultured ICM, fixed
5 days after immunosurgery. The open star marks the region equivalent to the
ectoplacental cavity of the normal embryo, while the closed star denotes the equivalent of the exocoelom, which is continuous with the proamniotic cavity. Bar is
100/im.
In all structures the outer cells resemble visceral endoderm of normal embryos,
with numerous microvilli, vacuoles, membrane-bound degenerative inclusions
and extensive endoplasmic reticulum. The basement membrane separating them
from the inner cells remains thin.
Cultured inner cell masses 5 days after immunosurgery
Between 3 and 5 days after immunosurgery a number of striking changes
take place in the internal organization of some of the larger cultured ICMs,
so that by day 5 about 25 % of them have a striking resemblance to the eggcylinder of normal C3H/He embryos at about 7-5 days p.c, with cells which
correspond in their morphology to embryonic and extra-embryonic ectoderm,
embryonic and extra-embryonic mesoderm and visceral and embryonic endoderm (Solter et al. 1970). These are classified as Type III structures (Fig. 3D,
6 A and B).
A possible intermediate stage between a Type II and Type III structure is
shown in Fig. 5. The closely packed, columnar epithelial layer presumably
corresponds to embryonic ectoderm, while the hemisphere of more rounded and
lighter staining epithelial cells resembles extra-embryonic ectoderm of normal
102
B. HOGAN AND R. TILLY
Fig. 7. Giant tiophoblast cells in culture. A batch of Type III cultured ICMs 5
days after immunosurgery was mechanically disrupted by sucking them into a
finely drawn out Pasteur pipette. The clumps of cells were then incubated in a 35 mm
tissue dish with DMEM plus 20 % HCS. After 2 days, scattered patches of cells
morphologically like trophoblast giant cells were present in the culture. They were
often associated with large clusters of rounded, less adherent cells (centre). Bar is
100 /tm.
embryos. The population of mesodermal-like cells seems to arise by delamination
from the embryonic ectoderm, over a localized region near the junction with
the presumed extra-embryonic ectoderm
In more advanced stages (Fig. 6 A, B) the mesoderm has apparently migrated
between the embryonic and extra-embryonic ectoderm and overlaying endoderm and has formed structures equivalent to the amnion and exocoelom of
the normal embryo. The embryonic ectoderm cells are bounded by a thin
basement membrane, except at the region of the 'primitive streak', and the
mesodermal cells always lie between this basement membrane and the endoderm
cells. The presumed extra-embryonic ectoderm is made up of both rounded
and more elongated cells which have few cytoplasmic organelles or microvilli,
and are joined along extensive areas of adjacent cytoplasm by junctional
complexes. In some embryos the hemisphere of extra-embryonic ectodermlike cells becomes rolled up into a flattened, chorion-like vesicle surrounding an
ectoplacental cavity. This is clearly shown in Fig. 6 A and B which are sections
at different levels through a single Type III cultured ICM. Of five other Type III
In vitro development of inner cell masses. I
103
structures sectioned every 20 /tm (see Methods), four resembled Fig. 6 A all
the way through, while one resembled Fig. 6B, with the chorion-like structure
only being found in some sections.
Several distinct kinds of outer endoderm cells are visible. Adjacent to the
junction between the embryonic and extra-embryonic ectoderm the endoderm
cells are very large, and the cytoplasm is filled with sheets of narrow endoplasmic
reticulum. At the base of the 'egg-cylinder', the endoderm cells are flattened
and have fewer microvilli and less endoplasmic reticulum. Both kinds of
endoderm closely resemble cells in equivalent positions in normal 7-5-day p.c.
embryos (Solter et al. 1970). Cells resembling parietal endoderm have never been
observed.
Further differentiation in vitro
When maintained in suspension culture the cultured ICMs do not develop
much beyond the stages shown in Figs. 3D and 6A and B. However, if the
embryo-like structures are mechanically disruped by pipetting through a finely
drawn out Pasteur pipette, clumps of cells attach to the substratum and after
3-5 days many different tissues are present, including beating muscle, red
blood cells and nerve. In addition, as shown in Fig. 7, large patches of cells
appear which closely resemble the giant trophoblast cells which develop when
extra-embryonic ectoderm from normal 9-day embryos is cultured in vitro
(Rossant & Ofer, 1977).
DISCUSSION
One of the striking features of these results is the high degree of organization
and differentiation which can be achieved in culture by individual inner cell
masses isolated immunosurgically from fully expanded blastocysts. On average
about 25 % of the inner cell masses develop within 5 days into structures which
closely resemble normal 7-5-day p.c. embryos. Further incubation has little
effect on the number of cultured ICMs showing this degree of development,
and many never progress beyond a simple hollow vesicle surrounded by endoderm. This failure to develop is probably a result of inadequate culture conditions, but although several attempts have been made to increase the yield of
Type III structures by using different culture media (e.g. Weymouth's medium,
RPM-1640, and increased levels of essential amino acids) these have not met
with any success.
A second striking feature is the presence in the most highly organized
cultured ICMs of cells which resemble the extra-embryonic ectoderm and
chorion of the normal 7-0-7-5-day p.c. C3H/He embryo (Fig. 6 A, B). In the
absence of specific intracellular biochemical markers, our tentative identification of these cells is based entirely on their morphology and tissue organization, and on the fact that cells resembling secondary trophoblast giant cells
104
B. HOGAN AND R. TILLY
are seen when mechanically disrupted Type III structures grow out on the
surface of the culture dish (Fig. 7). The extra-embryonic ectoderm-like cells
seem to develop from elongated cells which first make their appearance on about
day 3 of culture as part of the inner vesicle of Type II structures.
There is evidence from embryo reconstruction experiments, in which ICMs
isolated microsurgically from fully expanded 3-5-day p.c. mouse blastocysts
are injected into trophoblast vesicles or blastocysts of a different genotype
(Gardner, Papaionnou & Barton, 1973; Gardner & Papaionnou, 1975; Gardner
& Johnson, 1975), that the extra-embryonic ectoderm and chorion is normally
derived from polar trophoblast and not from the inner cell mass. In the light
of these embryo reconstitution experiments there are four possible explanations
for our results:
(1) In 25 % of the blastocysts the immunosurgery procedure failed to kill all
of the outer polar trophectoderm cells, so that one or two of them (possibly
with lower antigen concentration on the surface) remained with the inner cell
mass, and later, as a result of cell division, manifested themselves as extraembryonic ectoderm. While we cannot totally eliminate this possibility we feel
that it is very unlikely, for several reasons. Firstly, the embryos were incubated
in a concentration of antiserum seven times higher than the minimum concentration found to be effective in the immunosurgery procedure (see Methods).
Secondly, the concentration of complement used in these experiments was at
least ten times higher than the minimum required to lyse all the outer cells
previously exposed to a 1:30 dilution of antiserum. Finally, in a batch of
fourteen blastocysts fixed 30 min after exposure to complement and serially
sectioned at 1 jam, all of the outer trophectoderm cells were clearly dead, including
those above the inner cell mass. (See also fig. 2 in Hogan & Tilly, 1978). Using
other criteria, Handyside & Barton (1977) have also concluded that immunosurgery effectively kills all of the outer trophectoderm cells.
(2) At the stage when immunosurgery was performed, about 25 % of the
blastocysts may have begun to form ectoplacental cones, so that the polar
trophectoderm layer would be more than one cell thick above the inner cell
mass. Thus, even if all of the outer trophectoderm cells are killed, one or two
cells could remain with the inner cell mass and eventually give rise to extraembryonic ectoderm. Nevertheless, in all blastocysts which were sectioned
and examined in the electron microscope the polar trophectoderm appeared
to form a distinct monolayer over the inner cell mass (as noted also by Nadijcka
& Hillman (1974) for substage four blastocysts). However, the morphology of
the two cell types is very similar and we cannot exclude the possibility that
internalization of some polar trophectoderm had taken place before immunosurgery.
(3) The mature chorion of the normal embryo may be of dual origin, containing extra-embryonic ectoderm cells derived both from the polar trophectoderm
(ectoplacental cone) and from the inner cell mass. The embryo reconstruction
In vitro development of inner cell masses. I
105
experiments do not allow sufficient resolution to completely eliminate this
possibility.
(4) Finally, some cells in the inner cell mass may remain pluripotent and
able to give rise to trophoblast derivatives (extra-embryonic ectoderm) when
faced with an abnormal environment in which no polar trophectoderm is
present. This hypothesis is not incompatible with the results of the embryo
reconstruction experiments described above since in those experiments donor
trophectoderm tissue was always present.
At present, we consider either of the last two alternatives the most likely, but
cannot distinguish between them. One way of eliminating hypothesis (2),
however, is to study the in vitro development of inner cell masses isolated immunosurgically from blastocysts around 3-5 days/?.c, when it is very unlikely that
internalization of the polar trophoblast would have occurred. The results of
these experiments and a full discussion of the combined results are given in the
following paper (Hogan & Tilly, 1978).
We thank Gillian McArthur, James Vernon, Frank Fitzjohn and the staff of the SPF
Animal Breeding Unit for their skilled technical assistance.
REFERENCES
R. L. & JOHNSON, M. H. (1975). Investigation of cellular interaction and deployment in the early mammalian embryo using interspecific chimaeras between the rat and
mouse. In CIBA Foundation Symposium 29, 183-196. Amsterdam: Elsevier.
GARDNER, R. L. & PAPAIOANNOU, V. E. (1975). Differentiation in the trophectoderm and
inner cell mass. In The Early Development of Mammals (ed. M. Balls & A. E. Wild), pp.
107-132. Cambridge University Press.
GARDNER, R. L., PAPAIOANNOU, V. E. & BARTON, S. C. (1973). Origin of the ectoplacental
cone and secondary giant cells in mouse blastocysts reconstituted from isolated trophoblast and inner cell mass. /. Embryol. exp. Morph. 30, 561-572.
HANDYSIDE, A. H. & BARTON, S. C. (1977). Evaluation of the technique of immunosurgery
for the isolation of inner cell masses from mouse blastocysts. /. Embryol. exp. Morph. 37,
217-226.
HOGAN, B. & TILLY, R. (1978). In vitro development of inner cell masses iolated immunosurgically from mouse blastocysts. II. Inner cell masses from 3-5- to 4-0-day p.c. blastocysts.
/. Embryol. exp. Morph. 45, 107-121.
NADIJCKA, M. & HILLMAN, N. (1974). Ultrastructural studies of the mouse blastocyst
substages. /. Embryol. exp. Morph. 32, 675-695.
ROSSANT, J. & OFER, L. (1977). Properties of extra-embryonic ectoderm isolated from postimplantation mouse embryos. /. Embryol. exp. Morph. 39, 183-194.
SOLTER, D., DAMJANOV, I. & SKREB, N. (1970). Ultrastructure of mouse egg cylinder. Z.
Anat. EntwGesch. 132, 291-298.
SOLTER, D. & KNOWLES, B. B. (1975). Immunosurgery of mouse blastocyst. Proc. natn.
Acad. Sci. U.S.A. 72, 5099-5102.
GARDNER,
{Received 13 September 1977, revised 12 December 1977)